Fatigue is the most frequently reported symptom in patients with cancer. The causes of fatigue are multifactorial and include the disease itself, antineoplastic therapies, anemia, depression, and malnutrition. Our preliminary data suggest that micronutrient deficiencies, specifically carnitine deficiency, may be an important factor in fatigue. Carnitine plays a major role in energy metabolism. Systemic depletion is characterized by weight loss, fatigue, muscle weakness, decreased tolerance to metabolic stress, andcardiomyopathy. We found deficiency of the micronutrient carnitine in 17/27 patients with cancer. Symptoms of fatigue and functional status improved significantly in those patients who received oral L-carnitine supplementation. In addition, we observed similar results in 6/10 patients with end stage AIDS presenting with carnitine deficiency. Objectives: a) To determine the effect of L-carnitine therapy on symptoms of fatigue in patients with terminal cancer and serum carnitine deficiency, and b) to determine the effect of L-carnitine therapy on performance status, cognitive function, mood, quality of life, and motor activity in these patients. Study Design: We propose to conduct a Phase Il developmental randomized double-blind placebo-controlled study to determine the effect of the micronutrient L-carnitine on fatigue and other outcomes in patients with terminal cancer. A sample of 130 patients with terminal cancer, fatigue and serum carnitine deficiency will be recruited into the intervention study. At the first visit, the patients will receive a baseline assessment of fatigue, performance status, cognitive function, mood, and quality of life and motor activity. a) During weeks 1-4, patients will be randomized to receive L-carnitine, 2 g/day, placebo. b) During weeks 5-8, all patients will receive L-carnitine at a dose of 2 g/day, for a period of 4 weeks. Measures will be repeated at 48 hs, 2 weeks, 4 weeks and 8 weeks. The primary endpoint will be change in fatigue at 4 weeks. Analysis will evaluate group differences in the primary endpoint and other outcomes. An interim analysis will be done once 15 patients have been observed in each group at 4 weeks. Paired and unpaired data analyses between groups will be conducted with the assistance of a statistician.